Jesse Haviland

∙

2 followers

My goal, as a robotics researcher, is to enable robots to function consistently and resiliently in real-world environments. Real-world environments are ​​outside of controlled laboratory conditions, and tend to be dynamic, unstructured, and unknowable. My research is highly interdisciplinary, operating at the intersection of computer science, control, computer vision, and artificial intelligence. See more at jhavl.com.

I graduated with First Class Honours in a Bachelor of Software and Electrical Engineering in 2018 before starting my PhD. My PhD titled Control Strategies for Reactive Manipulation was completed in 2022 and supervised by Peter Corke. This research was undertaken at the Australian Centre for Robotic Vision and the QUT Centre for Robotics. I started as a Postdoctoral Research Fellow at the QUT Centre for Robotics and a Visiting Scientist at CSIRO in 2022. In 2023, I commenced as a Lecturer at QUT where I am teaching Foundations of Kinematics and Algorithms in Robotics and Introduction to Robotics.

Research Focus

I am interested in exploring how high level task planning, low level control theory and sensory input can enable robots to complete difficult and complex tasks while reacting to changes in the environment and task. This can be broken down into several topics:

- Utilising all of the joints in a robot be utilised to create fluid and smooth motion – more akin to human movement than robotic movement.

- Incorporating fast closed-loop motion controllers with robotic vision to enable robots to react to changes in the environment.

- Break a complex high level task into subtasks such that existing low level motion controllers can be utilised.

I am an active member of the international robotics community and have been an invited speaker at several workshops at IEEE International Conference on Robotics and Automation and the IEEE Conference on Intelligent Robots and Systems. I have also been an invited lecturer at several IEEE winter and summer schools. In 2023 I organised the workshop Taking Mobile Manipulators into the Real World held at Robotics: Science and Systems.

Open Source Software

Along with my collaborator Peter Corke, I am the co-creator of Python Robotics, an open-source ecosystem of Python packages for robotics research, education and industry. This ecosystem includes the very well known Robotics Toolbox for Python, a Python package for arm-type and mobile robots. Other packages support robotic vision, spatial mathematics, robot simulation, block diagram simulation, and graphing. Collectively, these packages have been downloaded over 500 000 times.